Use prisma internal field id value as a public resource identifier. Hello! I would like to implement a web application for a blog, where posts will be accessible via a URI, which will contain a unique identifier for each post. As example “http://mysite.com/posts/<database_post_id>”. Is the public internal database identifier correct as a public resource identifier? Is it bad idea?

Has anyone tried prisma with Apollo Federation? It should work just fine right?

yeah I am using it

hi. no `$fragment`s in `create*()`s?

i'd like to get to the nested object's creation `id`s

@joar Apollo Federation looks really good. I have a project I want to put it on. How do you like it?

Can anyone recommend a public graphql api (ideally with both a playground and subscription support) that I can use as a default in our boilerplate project?

I'm trying to figure out how to specify an earlier version of Prisma (1.30) when creating a Heroku Instance through the prisma cloud interface. I'm not seeing anywhere that you can change it. Do you have to create it with 1.34? I'm wanting to just get up and running on 1.30 for now and will figure out all the new

@relation(link: TABLE)

later

Can we create separate #heroku and #nexus channels?

Can someone help me with this? https://github.com/prisma/prisma2/issues/179

Anyone know how to get UUIDs working with Prisma2? doing just

@default(uuid()) @id @unique)

doesn't not do it. Getting error that I haven't included the UUID type in the main query

I have a question about how prisma batches queries - We have a schema where a user has multiple projects and is trying to fetch all of them with a query - I am printing out the PostgreSQL logs and it looks like prisma makes a separate database call per project, where i would expect the behavior to make one database call that returns both projects. I’ve attached the graphql query and the database log in the replies. Why is prisma not batching the query for projects on user in this case?

Hi everyone , Can anyone tell me how to write raw query ? i’m using prisma version 1.32.0 and not using docker. Thanks in advance.

hello, are there any general best practices for using prisma for microservices. the docs are unclear

there are no hooks for . pre-deploy . has anyone did some hacks to get it working

Hi im having an issue im calling the create mutation but getting the following error : "cannot return null for non-nullable type" HELPPPPPPPPPPPPPPPPPPPP

Hi, I hope this is the right place to ask this question, please let me know otherwise. I'm just getting started with prisma, looking for a production-ready alternative for our rethinkdb setup. Should I be looking into prisma 1 or 2? Website says 2 is still in 'preview', but I can't seem to find information on what exactly that means. Will it be a huge effort to migrate from 1 to 2, if I stick with 1 now and wait for 2 to become more stable? Thanks so much in advance

Is there any one who solve this issue? TS2590: Expression produces a union type that is too complex to represent. It was come from typescript version and nexus-prisma Here is github issue https://github.com/prisma/nexus-prisma/issues/284 Thank you.

does launch of prisma 2 means that prisma 1 wont be work on anymore like graphcool which is in maintenance phase ?

Has anybody used Nexus-Prisma with two prisma endpoints. I'm trying to add data from another from one endpoint to the other, but it doesn't recognize the types of the second endpoint.

hey! is there a way how to re-write

pageInfo

or

aggregate { count }

to return

totalCount

? thanks!

Hi there! We’re looking for an event sourcing layer for certain models in our database (need to access historic data). Are there any good packages availabel that allow us to do this?

two questions about deployments: I've got a node (apollo + prisma-client) app I want to deploy to a lambda. 1. Does it make sense to deploy prisma server (v1.x) to AWS Fargate? That seems like the best practice 2. I noticed prisma-client has a large number of deps, but I can't find anything about running

prisma generate

with some kind of

--production

flag. Are things like

flow

used in production?

I just got through the whole “Deploying your prisma app to AWS Fargate” process and I didn’t deploy what I thought I was. I didn’t realize that this tutorial deploys the portion that I essentially already had deployed via prisma.io. What I was really looking to do is instead deploy the prisma client/server that was connecting to Prisma.io. Is there a tutorial for that as well?

hey folks am I correct in thinking that

updateMany…

mutations still do not support nested mutations?

has anyone ever been able to scale prisma? we’ve encountered severe outages with <2k concurrent users. if you’ve solved this before, let me know, as i’m willing to pay for help at this point

@Teri Scaling prisma horizontally would require a message queue. There was a thread somewhere about it, and there is a lib you would have to add to your docker. I think its rabbitMQ if I am not mistaken. I have also encountered some scaling issues with prisma. There is a limitation set on Slick on the amount of 1000 tasks. You will get concurrent errors once you have reached that. I assume that this is what you are getting. I solved this a couple of things: 1.) Pagination ( Helps a lot ) 2.) It will be a good idea to have some sort of message queue in front of your service. Or even a load balancer if you plan on doing more than one instance. 3.) I also ended up using a cache layer ( Redis ) to limit the amount of requests being made. Then some data can feed out of Redis instead of asking prisma for data. At the end of the day it was a bunch of things that I had to do. It came down to limiting the requests and/or the amount of tasks being created per request in prisma

I hope this helps 🙂 All the best of luck

Hey guys, I have several fields of one type, and I have a filter for each of them on a frontend. My goal is to be able to filter items by one field if it’s selected on a frontend and other don’t and filter with multiple fields if they are selected on a frontend. The current problem is that when I pass data to a resolver from frontend it just sends a

null

value to a field and my filter on a backend is looking for items which has a

null

value in that specified field but I don’t want to filter by

null

In prisma2, how do I specify that I don't want to have a text type in Postgres for a String field but varchar?